find-skills-ja
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and analyzing SKILL.md files from the public skills.sh registry.
- Ingestion points: The agent is instructed to fetch and read the content of SKILL.md files from external URLs (GitHub raw or skills.sh) to perform safety checks.
- Boundary markers: While the skill instructs the agent to look for malicious commands, it lacks technical delimiters to separate untrusted content from the agent's instructions.
- Capability inventory: The skill can trigger the installation of third-party code via 'npx skills add' and pipe external output directly to the agent via 'npx skills use'.
- Sanitization: No automated filtering or sanitization of external content is specified beyond manual agent review.
- [COMMAND_EXECUTION]: The skill utilizes the 'skills' CLI to manage environment extensions.
- Evidence: Use of 'npx skills find', 'npx skills add', and 'npx skills ls' commands in SKILL.md.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the discovery and download of external agent code and metadata.
- Evidence: Instructions in SKILL.md Step 4 describe fetching remote SKILL.md content for review before recommending installation.
Audit Metadata